LD04 ch14 PDF

Fundamentals of Multimedia, Chapter 14
Chapter 14
MPEG Audio Compression
14.1 Psychoacoustics
14.2 MPEG Audio
14.3 Other Commercial Audio Codecs
14.4 The Future: MPEG-7 and MPEG-21
14.5 Further Exploration
1 Li & Drew c Prentice Hall 2003
14.1 Psychoacoustics
The range of human hearing is about 20 Hz to about 20 kHz
The frequency range of the voice is typically only from about
500 Hz to 4 kHz
The dynamic range, the ratio of the maximum sound ampli-
tude to the quietest sound that humans can hear, is on the
order of about 120 dB
Equal-Loudness Relations
Fletcher-Munson Curves
Equal loudness curves that display the relationship be-
tween perceived loudness (Phons, in dB) for a given
stimulus sound volume (Sound Pressure Level, also in
dB), as a function of frequency
Fig. 14.1 shows the ears perception of equal louness:
The bottom curve shows what level of pure tone stimulus
is required to produce the perception of a 10 dB sound
All the curves are arranged so that the perceived loudness
level gives the same loudness as for that loudness level of
a pure tone at 1 kHz
0
20
40
60
80
100
120
Equal Loudness Response Curves for the Human Ear
Frequency (kHz)
L
o
u
d
n
e
s
s

L
e
v
e
l

(
d
B
)
10dB
20dB
40dB
50dB
60dB
70dB
30dB
80dB
90dB
1.0 10.0
0.1
Fig. 14.1: Fletcher-Munson Curves
(re-measured by Robinson and Dadson)
Frequency Masking
Lossy audio data compression methods, such as MPEG/Audio
encoding, remove some sounds which are masked anyway
The general situation in regard to masking is as follows:
1. A lower tone can eectively mask (make us unable to
hear) a higher tone
2. The reverse is not true a higher tone does not mask a
lower tone well
3. The greater the power in the masking tone, the wider is
its inuence the broader the range of frequencies it can
mask.
4. As a consequence, if two tones are widely separated in
frequency then little masking occurs
Threshold of Hearing
A plot of the threshold of human hearing for a pure tone
10
2
10
3
10
4
10
0
10
20
30
40
50
60
Hz
d
B
Fig. 14.2: Threshold of human hearing, for pure tones
Threshold of Hearing (contd)
The threshold of hearing curve: if a sound is above the dB
level shown then the sound is audible
Turning up a tone so that it equals or surpasses the curve
means that we can then distinguish the sound
An approximate formula exists for this curve:
Threshold(f) = 3.64(f/1000)
0.8
6.5e
0.6(f/10003.3)
2
+ 10
3
(f/1000)
4
(14.1)
The threshold units are dB; the frequency for the origin
(0,0) in formula (14.1) is 2,000 Hz: Threshold(f) = 0 at
f =2 kHz
Frequency Masking Curves
Frequency masking is studied by playing a particular pure
tone, say 1 kHz again, at a loud volume, and determining how
this tone aects our ability to hear tones nearby in frequency
one would generate a 1 kHz masking tone, at a xed
sound level of 60 dB, and then raise the level of a nearby
tone, e.g., 1.1 kHz, until it is just audible
The threshold in Fig. 14.3 plots the audible level for a single
masking tone (1 kHz)
Fig. 14.4 shows how the plot changes if other masking tones
are used
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
10
0
10
20
30
40
50
60
70
Frequency (kHz)
d
B
Audible tone
Inaudible tone
Fig. 14.3: Eect on threshold for 1 kHz masking tone
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
10
0
10
20
30
40
50
60
70
Frequency (kHz)
d
B
1 4 8
Fig. 14.4: Eect of masking tone at three dierent frequencies
Critical Bands
Critical bandwidth represents the ears resolving power for
simultaneous tones or partials
At the low-frequency end, a critical band is less than
100 Hz wide, while for high frequencies the width can
be greater than 4 kHz
Experiments indicate that the critical bandwidth:
for masking frequencies < 500 Hz: remains approximately
constant in width ( about 100 Hz)
for masking frequencies > 500 Hz: increases approxi-
mately linearly with frequency
Table 14.1 25-Critical Bands and Bandwidth
Band # Lower Bound Center Upper Bound Bandwidth
(Hz) (Hz) (Hz) (Hz)
1 - 50 100 -
2 100 150 200 100
3 200 250 300 100
4 300 350 400 100
5 400 450 510 110
6 510 570 630 120
7 630 700 770 140
8 770 840 920 150
9 920 1000 1080 160
10 1080 1170 1270 190
11 1270 1370 1480 210
12 1480 1600 1720 240
Band # Lower Bound Center Upper Bound Bandwidth
(Hz) (Hz) (Hz) (Hz)
13 1720 1850 2000 280
14 2000 2150 2320 320
15 2320 2500 2700 380
16 2700 2900 3150 450
17 3150 3400 3700 550
18 3700 4000 4400 700
19 4400 4800 5300 900
20 5300 5800 6400 1100
21 6400 7000 7700 1300
22 7700 8500 9500 1800
23 9500 10500 12000 2500
24 12000 13500 15500 3500
25 15500 18775 22050 6550
Bark Unit
Bark unit is dened as the width of one critical band, for
any masking frequency
The idea of the Bark unit: every critical band width is roughly
equal in terms of Barks (refer to Fig. 14.5)
250 Hz
0
20
40
60
500 Hz 1 kHz 8 kHz 4 kHz 2 kHz
Critical band number (Bark)
d
B
0 25 20 15 10 5
80
Fig. 14.5: Eect of masking tones, expressed in Bark units
Conversion: Frequency & Critical Band Number
Conversion expressed in the Bark unit:
Critical band number (Bark) =
f/100, for f < 500 ,

9 + 4 log
2
(f/1000), for f 500 .
(14.2)
Another formula used for the Bark scale:
b = 13.0 arctan(0.76f) +3.5arctan(f
2
/56.25) (14.3)
where f is in kHz and b is in Barks (the same applies to all below)
The inverse equation:
f = [(exp(0.219b)/352)+0.1]b0.032exp[0.15(b5)
2
] (14.4)
The critical bandwidth (df) for a given center frequency f can also be
approximated by:
df = 25 + 75 [1 +1.4(f
2
)]
0.69
(14.5)
Temporal Masking
Phenomenon: any loud tone will cause the hearing receptors
in the inner ear to become saturated and require time to
recover
The following gures show the results of Masking experi-
ments:
100
Delay time (ms)
d
B
Test tone
Mask tone
60
40
20
1000 10 0 5
Fig. 14.6: The louder is the test tone, the shorter it takes for
our hearing to get over hearing the masking.
0
0.01
0.02
0.03
0
4
6
8
10
0
10
20
30
40
50
60
Frequency
Time
L
e
v
e
l

(
d
B
)
Tones below surface
are inaudible
Fig. 14.7: Eect of temporal and frequency maskings depending
on both time and closeness in frequency.
10
d
B
60
40
20
Delay time (ms)
100 0 5 50
Fig. 14.8: For a masking tone that is played for a longer time,
it takes longer before a test tone can be heard. Solid curve:
masking tone played for 200 msec; dashed curve: masking tone
played for 100 msec.
14.2 MPEG Audio
MPEG audio compression takes advantage of psychoa-
coustic models, constructing a large multi-dimensional lookup
table to transmit masked frequency components using fewer
bits
MPEG Audio Overview
1. Applies a lter bank to the input to break it into its fre-
quency components
2. In parallel, a psychoacoustic model is applied to the data
for bit allocation block
3. The number of bits allocated are used to quantize the
info from the lter bank providing the compression
MPEG Layers
MPEG audio oers three compatible layers :
Each succeeding layer able to understand the lower layers
Each succeeding layer oering more complexity in the psy-
choacoustic model and better compression for a given
level of audio quality
each succeeding layer, with increased compression eec-
tiveness, accompanied by extra delay
The objective of MPEG layers: a good tradeo between
quality and bit-rate
MPEG Layers (contd)
Layer 1 quality can be quite good provided a comparatively
high bit-rate is available
Digital Audio Tape typically uses Layer 1 at around 192 kbps
Layer 2 has more complexity; was proposed for use in Digital
Audio Broadcasting
Layer 3 (MP3) is most complex, and was originally aimed at
audio transmission over ISDN lines
Most of the complexity increase is at the encoder, not the
decoder accounting for the popularity of MP3 players
MPEG Audio Strategy
MPEG approach to compression relies on:
Quantization
Human auditory system is not accurate within the width
of a critical band (perceived loudness and audibility of a
frequency)
MPEG encoder employs a bank of lters to:
Analyze the frequency (spectral) components of the au-
dio signal by calculating a frequency transform of a win-
dow of signal values
Decompose the signal into subbands by using a bank of
lters (Layer 1 & 2: quadrature-mirror; Layer 3: adds
a DCT; psychoacoustic model: Fourier transform)
MPEG Audio Strategy (contd)
Frequency masking: by using a psychoacoustic model to
estimate the just noticeable noise level:
Encoder balances the masking behavior and the available
number of bits by discarding inaudible frequencies
Scaling quantization according to the sound level that is
left over, above masking levels
May take into account the actual width of the critical bands:
For practical purposes, audible frequencies are divided into
25 main critical bands (Table 14.1)
To keep simplicity, adopts a uniform width for all fre-
quency analysis lters, using 32 overlapping subbands
MPEG Audio Compression Algorithm
What to drop
Audio
(PCM)
input
Psychoacoustic
modeling
Bit allocation,
quantizing and
coding
Bitstream
formatting
Time to
frequency
transformation
Encoded
bitstream
Frequency
to time
transformation
Bitstream
unpacking
Frequency
sample
reconstruction
Decoded
PCM audio
Encoded
bitstream
Fig. 14.9: Basic MPEG Audio encoder and decoder.
Basic Algorithm (contd)
The algorithm proceeds by dividing the input into 32 fre-
quency subbands, via a lter bank
A linear operation taking 32 PCM samples, sampled in
time; output is 32 frequency coecients
In the Layer 1 encoder, the sets of 32 PCM values are rst
assembled into a set of 12 groups of 32s
an inherent time lag in the coder, equal to the time to
accumulate 384 (i.e., 1232) samples
Fig.14.11 shows how samples are organized
A Layer 2 or Layer 3, frame actually accumulates more
than 12 samples for each subband: a frame includes 1,152
samples
12
samples
Each subband filter produces 1 sample out
for every 32 samples in

Audio (PCM)
samples In
Subband filter 0
Subband filter 1
Subband filter 2
Subband filter 31
Layer 1
Frame
Layer 2 and Layer 3
Frame
12
samples
12
samples
12
samples
12
samples
12
samples
12
samples
12
samples
12
samples
12
samples
12
samples
12
samples
.

.

.
.

.

.
.

.

.
.

.

.
Fig. 14.11: MPEG Audio Frame Sizes
Bit Allocation Algorithm
Aim: ensure that all of the quantization noise is below the
masking thresholds
One common scheme:
For each subband, the psychoacoustic model calculates the Signal-
to-Mask Ratio (SMR)in dB
Then the Mask-to-Noise Ratio (MNR) is dened as the dierence
(as shown in Fig.14.12):
MNR
dB
SNR
dB
SMR
dB
(14.6)
The lowest MNR is determined, and the number of code-bits allocated
to this subband is incremented
Then a new estimate of the SNR is made, and the process iterates
until there are no more bits to allocate
Sound pressure
level (db)
Masker
Minimum
masking threshold
Neighboring
band
Critical band Neighboring
band
Bits allocated
to critical band
Frequency
m1
m+1
m
S
N
R
S
M
R
M
N
R
Fig. 14.12: MNR and SMR. A qualitative view of SNR, SMR and
MNR are shown, with one dominate masker and m bits allocated
to a particular critical band.
Mask calculations are performed in parallel with subband l-
tering, as in Fig. 4.13:
PCM
audio signal
Linear
quantizer
Bitstream
formatting
Filter bank:
32 subbands
1,024-point
FFT
Psychoacoustic
model
Coded audio
signal
Side-information
coding
Fig. 14.13: MPEG-1 Audio Layers 1 and 2.
Layer 2 of MPEG-1 Audio
Main dierence:
Three groups of 12 samples are encoded in each frame and
temporal masking is brought into play, as well as frequency
masking
Bit allocation is applied to window lengths of 36 samples
instead of 12
The resolution of the quantizers is increased from 15 bits
to 16
Advantage:
a single scaling factor can be used for all three groups
Layer 3 of MPEG-1 Audio
Main dierence:
Employs a similar lter bank to that used in Layer 2,
except using a set of lters with non-equal frequencies
Takes into account stereo redundancy
Uses Modied Discrete Cosine Transform (MDCT) ad-
dresses problems that the DCT has at boundaries of the
window used by overlapping frames by 50%:
F(u) = 2
N1
i=0
f(i) cos
2
N
i +
N/2 +1
2
(u +1/2)
, u = 0, .., N/2 1
(14.7)
PCM
audio signal
Filter bank:
32 subbands
1,024-point
FFT
Psychoacoustic
model
M-DCT
Nonuniform
quantization
Bitstream
formatting
Huffman
coding
Side-information
coding
Coded audio
signal
Fig 14.14: MPEG-Audio Layer 3 Coding.
Table 14.2 shows various achievable MP3 compression ratios:
Table 14.2: MP3 compression performance
Sound Quality Bandwidth Mode Compression
Ratio
Telephony 3.0 kHz Mono 96:1
Better than 4.5 kHz Mono 48:1
Short-wave
Better than 7.5 kHz Mono 24:1
AM radio
Similar to 11 kHz Stereo 26 - 24:1
FM radio
Near-CD 15 kHz Stereo 16:1
CD > 15 kHz Stereo 14 - 12:1
MPEG-2 AAC (Advanced Audio Coding)
The standard vehicle for DVDs:
Audio coding technology for the DVD-Audio Recordable
(DVD-AR) format, also adopted by XM Radio
Aimed at transparent sound reproduction for theaters
Can deliver this at 320 kbps for ve channels so that
sound can be played from 5 dierent directions: Left,
Right, Center, Left-Surround, and Right-Surround
Also capable of delivering high-quality stereo sound at bit-
rates below 128 kbps
MPEG-2 AAC (contd)
Support up to 48 channels, sampling rates between 8 kHz
and 96 kHz, and bit-rates up to 576 kbps per channel
Like MPEG-1, MPEG-2, supports three dierent proles,
but with a dierent purpose:
Main prole
Low Complexity(LC) prole
Scalable Sampling Rate (SSR) prole
MPEG-4 Audio
Integrates several dierent audio components into one stan-
dard: speech compression, perceptually based coders, text-
to-speech, and MIDI
MPEG-4 AAC (Advanced Audio Coding), is similar to the
MPEG-2 AAC standard, with some minor changes
Perceptual Coders
Incorporate a Perceptual Noise Substitution module
Include a Bit-Sliced Arithmetic Coding (BSAC) module
Also include a second perceptual audio coder, a vector-
quantization method entitled TwinVQ
MPEG-4 Audio (Contd)
Structured Coders
Takes Synthetic/Natural Hybrid Coding (SNHC) in or-
der to have very low bit-rate delivery an option
Objective: integrate both natural multimedia sequences,
both video and audio, with those arising synthetically
structured audio
Takes a toolbox approach and allows specication of
many such models.
E.g., Text-To-Speech (TTS) is an ultra-low bit-rate method,
and actually works, provided one need not care what the
speaker actually sounds like
14.3 Other Commercial Audio Codecs
Table 14.3 summarizes the target bit-rate range and main
features of other modern general audio codecs
Table 14.3: Comparison of audio coding systems
Codec Bit-rate Complexity Main
kbps/channel Application
Dolby AC-2 128-192 low (en-/decoder) p-to-p, cable
Dolby AC-3 32-640 low (decoder) HDTV, cable, DVD
Sony ATRAC 140 low (en-/decoder) minidisc
14.4 The Future: MPEG-7 and MPEG-21
Dierence from current standards:
MPEG-4 is aimed at compression using objects.
MPEG-7 is mainly aimed at search: How can we nd
objects, assuming that multimedia is indeed coded in terms
of objects
MPEG-7: A means of standardizing meta-data for au-
diovisual multimedia sequences meant to represent in-
formation about multimedia information
In terms of audio: facilitate the representation and search
for sound content. Example application supported by
MPEG-7: automatic speech recognition (ASR).
MPEG-21: Ongoing eort, aimed at driving a standard-
ization eort for a Multimedia Framework from a con-
sumers perspective, particularly interoperability
In terms of audio: support of this goal, using audio.
14.5 Further Exploration
Link to Further Exploration for Chapter 14.
In Chapter 14 the Further Exploration section of the text web-
site, a number of useful links are given:
Excellent collections of MPEG Audio and MP3 links.
The ocial MPEG Audio FAQ
MPEG-4 Audio implements Tools for Large Step Scala-
bility, An excellent reference is given by the Fraunhofer-
Gesellschaft research institute, MPEG 4 Audio Scalable Pro-
le.

LD04 ch14 PDF

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

LD04 ch14 PDF

Uploaded by

Copyright:

Available Formats

Fundamentals of Multimedia, Chapter 14

f/100, for f < 500 ,

You might also like